According to Reuters, Dutch digital mapping company TomTom announced an exciting new partnership with tech giant Microsoft on Tuesday to develop an artificial intelligence (AI) powered conversational assistant for vehicles. This assistant utilizes Microsoft’s advanced Azure OpenAI service and allows drivers to “converse naturally with their vehicles,” interacting through voice with features like infotainment, navigation, and vehicle commands.

As vehicles become more advanced, automakers have been focused on integrating smart assistant technology into dashboards. Voice-powered assistants like Alexa and Siri have become ubiquitous in our homes and phones, so it’s a logical next step to bring similar functionality into our cars to handle tasks while driving. This allows drivers to access useful features while keeping their eyes on the road.

Microsoft Brings the AI Power

To create an assistant that can handle natural conversations, vast amounts of data and immense computing power is needed. This is where Microsoft lent its capabilities through Azure OpenAI Service, providing access to the advanced language models created by leading AI research company OpenAI (the maker of ChatGPT). The cutting-edge AI can understand context and hold conversations – perfect for an in-vehicle assistant.

While TomTom could have integrated an existing voice assistant like Alexa, it chose instead to custom-build one explicitly optimized for the automotive environment. As a digital mapping pioneer with over 25 years in navigation and location technology, TomTom has the specialized expertise to tailor the AI assistant to handle driving-related requests through voice interactions.

The conversational assistant can provide quick access to useful in-car features by just using your voice. Ask for directions, queue up your favorite playlist, adjust vehicle settings, and more without ever taking your eyes off the road. This promises a safer and more convenient driving experience. TomTom can integrate the assistant with vehicle infotainment systems across multiple automotive brands.

This launch builds on an ongoing multi-year partnership between TomTom and Microsoft focused on pushing AI and location technology innovations. TomTom began utilizing Azure cloud tools in 2016 to power its maps and navigation products. With this latest AI assistant, TomTom continues to discover new ways Microsoft’s leading AI capabilities can enhance its offerings.

Featured Image Credit: Erik Mclean; Pexels

The post TomTom brings AI assistant to cars with Microsft’s help appeared first on ReadWrite.

Instagram rolled out its new AI-generated background editing tool to U.S.-based users. The feature, powered by generative AI, allows users to transform the backgrounds in their Instagram stories and photos through text prompts.

The background editor is being spearheaded by Ahmad Al-Dahle, Meta’s lead for generative AI. In an announcement post on Threads, Al-Dahle said the tool will enable users to prompt changes to their story backgrounds, opening up creative new possibilities.

“With backdrop, you can reimagine your image’s background with just a few taps and a prompt like ‘chased by dinosaurs’ or ‘surrounded by puppies’ to create an entirely new image for your Story. Tap the button for a backdrop at the top of a new Story to get started,’” Al-Dahle wrote. Users can also write their own custom prompts.

Now you can try the Instagram “Try it” stickers

Once posted to stories, the AI-generated backgrounds will feature a “Try it” sticker so other users can test out the same text prompts on their own images.

The launch builds on Instagram’s growing suite of AI offerings, including creative tools already available on Snapchat and other Meta platforms like Messenger and WhatsApp. Earlier this month, Meta rolled out a similar AI image generator called Imagine, while Snapchat unveiled paid features enabling users to create fantasy-themed photos.

As AI image generation technology advances, social platforms are racing to bake the functionality into their apps. Instagram hopes features like the background editor will set it apart, though ethical concerns around deepfakes continue to swirl around AI-powered creative tools.

Generative AI refers to a type of artificial intelligence capable of creating new content on its own instead of merely analyzing existing content. The “generative” part refers to the fact that these AI systems can generate brand-new images, text, audio, video, and other data that have never existed before.

Generative AI relies on deep learning algorithms, specifically on a subset of AI approaches known as neural networks. To learn patterns and concepts, these neural nets are trained on vast datasets – sometimes containing millions or billions of examples. Once the generative AI model has been sufficiently trained, it can use its learnings to produce novel, original creations based on what it has detected within its training data.

Featured Image: Photo by Pixabay; Pexels

The post Instagram rolls out AI-powered background editor appeared first on ReadWrite.

OpenAI, the leading company behind the viral ChatGPT AI chatbot, announced a new partnership with Axel Springer, one of the world’s biggest news publishers.

The collaboration aims to enhance ChatGPT’s knowledge and capabilities around current events and real-time information.

Under the deal, Axel Springer will provide summaries and extracts from its reporting to OpenAI, including articles from major titles like Politico, Business Insider, BILD, and WELT. OpenAI plans to use this content to train its AI systems, as well as surface relevant Axel Springer articles to ChatGPT users when they ask questions tied to recent news.

The summaries provided to ChatGPT will contain attribution back to the full Axel Springer articles, essentially providing publicity and traffic for the publisher. Users will also be able to ask ChatGPT questions that reference the information contained across the German media conglomerate’s journalism brands.

OpenAI and Axel Springer bring news to ChatGPT

The collaboration appears aimed at addressing one of the main criticisms of ChatGPT — its lack of up-to-date knowledge on current events. Services like xAI’s new Grok chatbot boast more real-time intelligence by tapping social media data streams. By licensing recent and archival content from major newsrooms, OpenAI hopes to match and exceed these capabilities.

The Springer deal builds on past OpenAI partnerships with the Associated Press and American Journalism Project to train AI models using journalism. As OpenAI continues to face questions around copyright infringement for scraping content without consent, licensing agreements with major publishers provide legal coverage — but also draw criticism around fairness and compensation for smaller sources.

While the partnership provides clear financial value to Springer, questions remain whether it’s ethical for OpenAI to strike such deals with large publishers while scraping content from independent creators and denying compensation. Striking the right balance around sourcing training data continues to be a complex challenge for even the most well-resourced AI companies.

The post OpenAI partners with Axel Springer in push for real-time content appeared first on ReadWrite.

The world’s most popular generative artificial intelligence (AI) is getting “lazy” as the winter draws in – that’s the claim from some astute ChatGPT users.

According to a recent ArsTechnica report in late November, users of ChatGPT, the AI chatbot powered by OpenAI’s natural language model GPT-4, began noticing something strange. In response to certain requests, GPT-4 was refusing to complete tasks or providing simplified “lazy” answers instead of the typically detailed responses.

OpenAI acknowledged the issue but claimed they did not intentionally update the model. Some now speculate this laziness may be an unintended consequence of GPT-4 mimicking seasonal human behavior changes.

Dubbed the “winter break hypothesis,” the theory suggests that because GPT-4 is fed the current date, it has learned from its vast training data that people tend to wrap up big projects and slow down in December. Researchers are urgently investigating whether this seemingly absurd idea holds weight. The fact it’s being taken seriously underscores the unpredictable and human-like nature of large language models (LLMs) like GPT-4.

On November 24th, a Reddit user reported asking GPT-4 to populate a large CSV file, but it only provided one entry as a template. On December 1st, OpenAI’s Will Depue confirmed awareness of “laziness issues” related to “over-refusals” and committed to fixing them.

Some argue GPT-4 was always sporadically “lazy,” and recent observations are merely confirmation bias. However, the timing of users noticing more refusals after the November 11th update to GPT-4 Turbo is interesting if coincidental and some assumed it was a new method for OpenAI to save on computing.

Entertaining the “Winter Break” theory

On December 9, developer Rob Lynch found GPT-4 generated 4,086 characters when given a December date prompt versus 4,298 for a May date. Although AI researcher Ian Arawjo couldn’t reproduce Lynch’s results to a statistically significant degree, the subjective nature of sampling bias with LLMs makes reproducibility notoriously difficult. As researchers rush to investigate, the theory continues intriguing the AI community.

Geoffrey Litt of Anthropic, Claude’s creator, called it “the funniest theory ever,” yet admitted it’s challenging to rule out given all the weird ways LLMs react to human-style prompting and encouragement, as shown by the increasingly weird prompts. For example, research shows GPT models produce improved math scores when told to “take a deep breath,” while the promise of a “tip” lengthens completions. The lack of transparency around potential changes to GPT-4 makes even unlikely theories worth exploring.

This episode demonstrates the unpredictability of large language models and the new methodologies required to understand their ever-emergent capabilities and limitations. It also shows the global collaboration underway to urgently assess AI advances that impact society. Finally, it’s a reminder that today’s LLMs still require extensive supervision and testing before being responsibly deployed in real-world applications.

The “winter break hypothesis” behind GPT-4’s apparent seasonal laziness may prove false or offer new insights that improve future iterations. Either way, this curious case exemplifies the strangely anthropomorphic nature of AI systems and the priority of understanding risks alongside pursuing rapid innovations.

Featured Image: Pexels

The post A look at the curious ‘winter break’ behavior of ChatGPT-4 appeared first on ReadWrite.

Online video platform Rumble is cracking down on Media Matters with a new lawsuit targeting people connected to the left-leaning website.

The suit from the video-sharing platform, popular with free speech advocates, alleges the non-profit media watchdog group engaged in misleading and unethical practices to damage its business.

The lawsuit filed on Nov 29, centers around Media Matters’ alleged strategy of repeatedly refreshing controversial videos on sites it aims to take down to capture major brand advertisements running adjacent to the content.

According to Rumble’s recent court filings, two individuals who run ‘Check my ads’ created fake profiles, sought out objectionable user-uploaded videos, and refreshed the pages multiple times until they could capture screenshots showing ads from high-profile brands next to the offending content. The data they collected was then used by Media Matters to call on big brands to avoid Rumble.

THREAD: Rumble is suing two individual Plaintiffs who run "Check my Ads" in connection with Media Matters. They are seeking punitive damages and an injunction to stop their conduct, among other things.

This will be my ongoing thread analyzing this lawsuit. Please bookmark it,… pic.twitter.com/QrKnx5w6JQ

— Tracy Beanz (@tracybeanz) December 12, 2023

Rumble said this strategy caused some brands, including Netflix, to pull their advertising spending on its platform after Media Matters showed them the controversially placed ads. The video platform called the tactics “unethical” and said the two defendants Nandini Jammi and Claire Atkin deliberately tried to damage sites it deemed controversial by orchestrating the ad placements.

Rumble’s legal action comes on the heels of the group’s parent company, X Corp, filing what owner CEO Elon Musk called a “thermonuclear” lawsuit against Media Matters last month. X Corp alleged Media Matters executed a similar “refresh” tactic on its platform to capture brand ads next to antisemitic content.

The split second court opens on Monday, X Corp will be filing a thermonuclear lawsuit against Media Matters and ALL those who colluded in this fraudulent attack on our company pic.twitter.com/55vl7PspaQ

— Elon Musk (@elonmusk) November 18, 2023

The lawsuit claims Media Matters then showed the screenshots to companies like Apple and IBM to prompt them to suspend their advertising partnerships with X Corp.

X Corp said Media Matters’ actions suppressed its free speech rights. Legal experts project X Corp may seek billions of dollars in damages from the media group.

Between the high-profile lawsuits from internet media giants Rumble and X Corp, Media Matters faces growing scrutiny and liability over its controversial editorial tactics. The non-profit will need to answer for its allegedly unethical practices in court in the months ahead.

Media Matters fights back

That’s not the end of the lawsuits. Media Matters is suing Texas Attorney General Ken Paxton after he opened an investigation into their reporting of  X.

In the lawsuit, the non-profit alleges that Mr Paxton’s “retaliatory” investigation is unlawful and violates its First Amendment rights.

Lawyers for Media Matters said: “The chill imposed by his retaliatory scheme injures Plaintiffs’ ability to investigate and publish news stories and further chills their ability to participate in a robust public discussion around political extremism on the X platform”.

In an interview with Reuters after Musk filed his lawsuit, Angelo Carusone, the Media Matters president, said the non-profit’s findings flew in the face of X’s statements that it had introduced safety protections to prevent ads from appearing next to harmful content.

“If you search for white nationalist content, there are ads flourishing. The system they say exists is not operating as such,” he said.

Featured Image: Pexels

The post Rumble sues two people connected to Media Matters on heels of X suit appeared first on ReadWrite.

In November 2022, the cybersecurity group vx-underground claimed on social media that unknown hackers had breached the now-defunct Bitcoin ATM provider Coin Cloud, according to a recent TechCrunch report. According to vx-underground, the hackers said they stole 70,000 customer photos and the personal data of 300,000 customers. This sensitive personal information allegedly included social security numbers, dates of birth, full names, email addresses, phone numbers, occupations, physical addresses, and more.

However, over a month later, many questions remain unanswered about this alleged hack. Nobody has publicly taken credit for the reported breach and details remain scarce, even to Coin Cloud’s new owner, Bitcoin ATM.

Coin Cloud filed for bankruptcy in February 2022 after operating thousands of cryptocurrency ATMs across the U.S. and Brazil. In July, the assets of Coin Cloud were acquired by Bitcoin ATM, a company owned by Andrew Barnard, who also owns another crypto ATM provider, Bitstop.

The company launched an investigation into the Coin Cloud system

Barnard told TechCrunch that upon learning of the vx-underground claims, his company launched an investigation into the Coin Cloud systems but found no concrete evidence of a recent breach. Barnard believes outdated Coin Cloud systems were breached multiple times in the past by various actors. He assumes the allegedly stolen customer data is from one of those previous attacks.

A former Coin Cloud employee we spoke to raised concerns about Coin Cloud’s poor security practices during its operational years before bankruptcy. They claim Coin Cloud had no dedicated security team, stored customer data in unencrypted plaintext, and was hacked at least once in 2022.

WithCoin Cloud’s systems and infrastructure now defunct after bankruptcy proceedings, Barnard admits it’s virtually impossible to determine precisely when or how the reported hack occurred. The culprits behind the alleged data breach remain shrouded in uncertainty.

While the vx-underground claims have not been definitively proven, they raise troubling questions about how the sensitive personal information of hundreds of thousands of Coin Cloud customers may have been compromised – and whether new owner Bitcoin ATM can prevent further data leaks. For now, the true scale and impact of the reported Coin Cloud hack continue to mystify.

Featured Image: Photo by Tima Miroshnichenko; Pexels

The post Coin Cloud bankruptcy leaves customers at risk appeared first on ReadWrite.

El Salvador’s ambitious plan to issue $1 billion in sovereign “Volcano Bonds” backed by Bitcoin has cleared a significant obstacle, bringing the pioneering offering closer to fruition, according to a recent Coindesk report. The country’s nascent Digital Assets Commission (CNAD) has granted regulatory approval for the bonds issued in early 2024 through Bitfinex Securities.

The approval represents a watershed moment for the experimental bonds first unveiled by President Nayib Bukele in 2021 following his government’s historic adoption of Bitcoin as legal tender. Enshrined in legislation passed by a vast parliamentary majority, the bonds are intended to allow El Salvador to tap financial markets to build a geothermal-powered Bitcoin mining ecosystem.

Initially slated for issuance in 2022, complications surrounding El Salvador’s negotiations with the IMF and designing a unique regulatory framework for the asset-backed offering delayed the bonds. Receiving the green light from CNAD, established in 2021 to govern El Salvador’s crypto industry, is seen as an affirmation that the country has managed to construct a robust architecture for crypto financial innovation.

El Salvador is positioning itself at the bleeding edge of digital asset adoption.

Positioning itself at the bleeding edge of digital asset adoption, El Salvador is moving forward with the bonds despite Bitcoin’s price volatility and the collapse of FTX demonstrating crypto’s risks. The Bukele administration views pioneering government adoption as a differentiator that enhances El Salvador’s investment attractiveness.

If successfully executed, the value of the bonds will be hedged to price stability while offering an attractive 6.5% coupon paid out in US dollars. The underlying bitcoin will be held in a transparent public wallet open to scrutiny, with El Salvador legally bound to redeem the bonds in traditional currency.

The issuance still faces challenges, like drumming up $1 billion in investment for an unproven crypto instrument issued by a developing economy. However, by clearing regulatory barriers, the Volcano Bonds have taken a major leap toward introducing the globe’s inaugural Bitcoin-backed sovereign debt.

Featured Image: Pexels

The post El Salvador Bitcoin bonds clear regulatory hurdle appeared first on ReadWrite.

In his first public appearance since regaining control of artificial intelligence (AI) startup OpenAI last week, co-founder and CEO Sam Altman spoke in defense of the company’s exploratory approach to developing AI technology, according to a recent Bloomberg report.

Addressing a conference hosted by rights organization Operation Hope in Atlanta on Monday, Altman stated “It’s going to lift the world up,” referring to AI’s potential, but stressed the importance of pushing boundaries safely to realize its benefits across healthcare, education, and more.

The comments come on the heels of recent turmoil in OpenAI’s leadership after Altman was briefly ousted as CEO in what he called a “coup attempt.” Concerns over the speed of development on OpenAI products like ChatGPT sparked the coup, stunning Silicon Valley. However, after investor and employee backlash, Altman was reinstated days later and most previous board members exited.

Altman acknowledged anxieties around AI being misused, saying “you don’t need much imagination because we grew up with that in the media.” But he defended OpenAI’s open, deploy-focused approach as key to developing the technology safely, rather than in secret.

Details around OpenAI’s new board and governance plans were not provided, and signs of ongoing differences with co-founder Ilya Sutskever remain after his opposition to Altman’s return.

The recent events have amplified scrutiny on AI advancements enabled by OpenAI that could reshape industries but also carry risks around military uses and other dangers according to experts.

Altman stated that “people have a lot of anxiety, and I get that,” referring to current societal apprehensions about advanced AI. He announced plans to co-chair a new AI ethics council with conference host John Hope Bryant as part of efforts to responsibly shape the emerging technology.

Featured Image: Pexels

The post OpenAI’s Sam Altman announces new AI ethics council appeared first on ReadWrite.

Shiprocket, an Indian logistics and shipping software startup, is currently in funding talks with venture capital firm Tribe Capital. According to an anonymous source familiar with the matter cited in a recent TechCrunch report, Tribe Capital is deliberating leading a $75-100 million investment into Shiprocket.

The funding talks are said to be ongoing, and the final terms may change, two people close to the discussions stated under the condition of anonymity due to the private nature of the talks. Both Tribe Capital and Shiprocket declined to comment when asked about the funding rumors.

New Delhi-based Shiprocket operates an e-commerce-focused logistics and shipping software solution targeted at courier services. Its platform utilizes a machine learning-powered data engine to recommend optimal couriers for businesses and handle tasks like printing shipping labels and tracking orders through a centralized interface. This enables companies to efficiently manage shipping and returns.

Shiprocket’s growing list of backers

Shiprocket already counts prominent investors like Temasek and Indian food delivery service Zomato among its backers. According to an internal company document reviewed by TechCrunch, Shiprocket is aiming to become IPO-ready within the next 12-18 months. As of September 2022, its annualized revenue run rate exceeded $165 million.

The company currently works with over 250,000 merchants across India who process around 200 million transactions annually through the platform. Shiprocket is planning to expand its financial offerings for merchants, including services like financing options and buy now, pay later for consumers.

The internal document projects Shiprocket’s revenue run rate to reach around $500 million by the end of 2025. Another anonymous source revealed plans to grow Shiprocket’s cross-border shipping capabilities and checkout offerings as well.

As an existing investor, Tribe Capital seems intent on increasing its India investments. In an interview earlier this year, Tribe Capital’s leader Arjun Sethi told Economic Times that the firm seeks to raise a new $250 million India-focused fund.

The talks come during a funding slowdown across Indian startups. 2023 has seen just $7 billion invested into Indian startups year-to-date according to Tracxn, an over 70% year-over-year decline from 2022 levels. Late-stage mega-rounds above $100 million are down almost 70% compared to last year.

Featured Image: Pexels

The post Tribe Capital in $100M funding talks with India’s Shiprocket: reports appeared first on ReadWrite.

Malta-based slots and casino games supplier Wazdan has secured a pivotal foothold in the Spanish igaming market through a new content distribution partnership with local operator Gran Madrid Casino Online, according to a recent Gambling Insider report.

The agreement, facilitated via Light & Wonder’s aggregation platform, marks Wazdan’s first foray into Spain. It also continues an aggressive growth strategy this year that has targeted locally-regulated markets across Europe and Latin America.

Spain was one of the first markets to introduce a legal framework for online gambling in Europe. However, complex licensing processes have led to slow expansion amongst international suppliers. By leveraging Gran Madrid’s existing local expertise, Wazdan sidesteps legislative barriers but gains access to an igaming sector expected to be worth $850 million by 2024.

The potential rewards are evidenced by Gran Madrid’s faith in player demand for new gaming products. Country Manager José Félix Herreros suggested as much in a statement, saying “We highly value this strategically important collaboration” to bring “unique gaming experiences” to Spanish customers.

For Wazdan, the move builds on previous deals with European heavyweights Kindred Group and Netwin Italia as the company targets growth through regulated markets. Kindred granted Wazdan access to major brands like Unibet and MariaCasino across territories like the UK, Malta and the Netherlands.

The Netwin partnership brought Wazdan’s portfolio of 140+ slots and table games to Italian operator sites for the first time in November.

Chief Commercial Officer Andrzej Hyla called landing Gran Madrid a “pivotal moment” in Wazdan’s Spanish strategy. With the complex licensing addressed, Hyla and his commercial team can now leverage Gran Madrid’s marketing apparatus and customer base to popularize Wazdan titles like Larry the Leprechaun, Great Book of Magic Deluxe and Lucky 9.

The chance for new player acquisition comes at an opportune time for Wazdan, 2023 has seen the company expand through regulated markets in Europe and make its first move into Latin America through recent Colombia and Mexico deals.

Featured image: Pexels

The post Wazdan forges key Spanish expansion through Gran Madrid deal appeared first on ReadWrite.

A new report from semiconductor research organization TechInsights reveals major advances in radio-frequency (RF) chip design and engineering in Huawei’s latest flagship smartphone, the Mate 60 Pro.

The device shows progress towards using domestically-produced components rather than relying on traditional American suppliers like Skyworks and Qorvo.

The Mate 60 Pro utilizes RF switches from Chinese company Maxscend Microelectronics and power amplifier modules from Beijing OnMicro Electronics. Previously, phones from Huawei and other Chinese manufacturers depended on equivalent parts from established U.S. leaders in RF technology.

“The architecture is tailored for the Chinese industry,” wrote TechInsights in their report released Monday. “While the Mate 60 Pro shows China narrowing the technological gap with Western competitors, it hasn’t completely closed it in 2023. However, given the Chinese government’s focus on reducing dependence on Western technology, this gap is expected to continue narrowing.”

The innovations indicated in the Mate 60 Pro prove that, even under U.S. trade restrictions, the Chinese tech giant can design cutting-edge mobile devices without American help.

Huawei Mate 60 Pro catching up with Western smartphones

This revelation comes on the heels of August’s launch of Huawei’s new 7-nanometer applications processor inside the Mate 60 Pro. Produced by Semiconductor Manufacturing International Corp. (SMIC) in Shanghai, it matched the complexity of equivalent system-on-chips from American giant Qualcomm.

An ongoing U.S. investigation into SMIC highlights authorities’ concerns about China’s rapid ascension in semiconductor technology. But Huawei’s persistence in fostering domestic alternatives for supply-chain constraints demonstrates the electronics giant’s autonomous capabilities regardless of trade barriers.

With sophisticated home-grown processors and now RF circuitry to match, the Mate 60 Pro epitomizes Huawei’s strides towards self-reliance. TechInsights’ analysis proves the company can stand toe-to-toe with smartphones utilizing even the most sophisticated American components. Huawei is rapidly realizing its vision of cutting-edge, made-in-China mobile devices.

The report follows recent reports that Chinese semiconductor factory Semiconductor Manufacturing International Corporation (SMIC) may have surpassed the 5nm process barrier despite U.S. restrictions — by developing a cutting-edge chip for Huawei.

Featured image: Huawei.com

The post Huawei Mate 60 Pro shows China ‘narrowing tech gap’ with US appeared first on ReadWrite.

Apps and websites that use artificial intelligence to digitally remove clothing from women’s photos without consent are skyrocketing in popularity, according to a Time report.

Researchers at the social network analysis firm Graphika found that in September alone, 24 million people visited so-called “nudify” or “undressing” services that apply AI algorithms to actual photos of clothed women to generate fake nude images.

The startling growth corresponds with the open-sourcing of several new AI diffusion models capable of creating strikingly realistic fake nude images. Whereas previous AI-altered images tended to be blurry or unrealistic, these new models allow amateurs to generate convincingly real nude photos without sophisticated training.

Marketing for these services often occurs openly on platforms like Reddit and Telegram, with some ads even suggesting users send the fake nudes back to the victim herself. One Nudify app pays for sponsored YouTube content and ranks first in Google searches despite the practice of violating most sites’ policies. Still, the services operate largely unchecked.

Deepfakes are dangerous, especially for non-consensual pornography

Experts warn the proliferation of DIY deepfake nudes represents a dangerous new phase in non-consensual pornography. “We are seeing more and more of this being done by ordinary people with ordinary targets,” said Eva Galperin of the Electronic Frontier Foundation, noting a rise in cases among high schoolers.

Still, many victims never even discover such images exist—those who do often struggle to get law enforcement to investigate or lack funds to pursue legal action.

Currently, federal law does not explicitly ban deepfake pornography, only the generation of faked child sexual abuse material. In November, a North Carolina psychiatrist received 40 years in prison – the first conviction of its kind – for using AI undressing apps on underage patient photos.

In response to the trend, TikTok has blocked search terms related to nudify apps and Meta began blocking related keywords on its platforms as well. Still, analysts say much more awareness and action is needed around non-consensual AI porn.

The apps treat women’s bodies as raw material without consent, running roughshod over privacy rights in pursuit of profit. “You can create something that actually looks realistic,” said Santiago Lakatos. Indeed, for victims, that is precisely the danger.

Featured Image Credit: Photo by Markus Winkler; Pexels

The post AI undressing deepfake nude generation apps soar in popularity appeared first on ReadWrite.

Nintendo has dealt a major blow to gaming fans by announcing the cancellation of its highly-anticipated Nintendo Live 2024 event scheduled for January in Tokyo, Japan, according to a recent Dexerto report. The extreme move came after multiple “disturbing and specific” threats of violence were made against both Nintendo employees and fans planning to attend the event.

Nintendo Live events have become legendary in the gaming community, offering die-hard Nintendo fans hands-on gameplay with new title releases, gaming tournaments, developer discussions, and exclusive Nintendo swag. Over 100,000 attendees were expected at Nintendo Live 2024 Tokyo which was set to feature high-stakes tournaments for hit games like Splatoon 3 and Mario Kart 8 Deluxe.

Nintendo says they regret canceling the event

In a translated statement, Nintendo expressed regret in having to cancel an event fans had waited years for but said “detailed and credible threats” left them no choice but to prioritize safety over the event. They did not disclose the specifics of the threats. Online discussion boards have lit up with fans expressing everything from disappointment to outright rage.

Though gaming events have dealt with threats before related to cheating disputes or match-fixing controversies, threats against fans are extremely rare, especially with violence. So far it is unclear if the threats are related to internal Nintendo conflicts or profit disputes with partner developers or if the company just found itself in the middle of larger sociopolitical turmoil crossfire.

Along with canceling Nintendo Live altogether, Nintendo also announced they have postponed the high-profile Splatoon 3 and Mario Kart tournaments indefinitely, with no details on if or when they may be rescheduled. Given the extreme step of canceling Nintendo Live altogether, the company appears to be taking the threats extremely seriously.

The post Nintendo live Tokyo canceled over violent threats appeared first on ReadWrite.

A team of researchers has uncovered a new vulnerability, dubbed “AutoSpill,” affecting several popular password managers on Android devices, according to a recent TechCrunch report. The vulnerability allows malicious apps to access users’ sensitive login credentials stored in password managers through exploiting the autofill function in Android’s WebView component.

According to Ankit Gangwal, Shubham Singh, and Abhijeet Srivastava from IIIT Hyderabad in India, the vulnerability works by tricking password managers into auto-filling credentials into an app’s native text fields when the app shows a login page through a WebView instead of launching an external web browser.

“Even without phishing, any malicious app that asks you to log in via another site, like Google or Facebook, can automatically access sensitive information,” explained Gangwal.

The researchers tested AutoSpill against leading password managers like 1Password, LastPass, Keeper, and Enpass on fully updated Android devices. They found most apps vulnerable even with JavaScript injection protections enabled. With JavaScript enabled, all tested password managers were susceptible.

The implications are serious considering the popularity of password managers to store sensitive login information for various online services. If exploited, the vulnerability provides malicious apps easy access to a treasure trove of usernames and passwords.

Gangwal disclosed the findings to both Google and the affected password manager developers. 1Password has acknowledged the vulnerability and claims to have identified a fix that will land in a future update. LastPass has also put mitigations in place. Other vendors have yet to publicly comment or confirm plans to address the issue.

The researcher team says they are still investigating whether a similar attack is possible on iOS devices. For now, Android users should be cautious about entering credentials into native app text fields, even if prompted by a WebView-based login page. As always, only install apps from trusted sources like the Google Play Store.

The AutoSpill research highlights the rising challenge of securing credentials across the modern app ecosystem. As login systems increasingly transition from web to in-app embedded browser flows, vulnerabilities like this demonstrate many risks are still being uncovered.

The post Many password managers vulnerable to AutoSpill attack on Android appeared first on ReadWrite.

Brussels is on the cusp of approving unprecedented new regulations on artificial intelligence (AI) systems, establishing the most comprehensive controls on AI yet seen in the Western world, according to a recent Bloomberg report.

In lengthy negotiations that stretched late into Wednesday night, delegates from the European Commission, Parliament, and 27 member states reached a tentative compromise on rules that would govern AI systems like chatbots and generative AI tools such as OpenAI’s ChatGPT and Google’s Bard. Sources close to the closed-door talks say a formal deal is expected early Thursday morning.

The new AI Act would be the first major regulation of AI systems outside of Asia and would fill a notable void left by inaction from the U.S. Congress. Technology policy experts see the rules as likely setting an influential precedent for AI governance globally.

Strict new controls for Generative AI

Among the most complex negotiations have centered around specific requirements for generative AI systems — AI tools capable of producing original text, images, audio, and video on command. These systems include chatbots like ChatGPT and creative tools like DALL-E 2.

The tentative deal would mandate that developers of generative AI maintain detailed documentation on how models are trained and what copyrighted source material is utilized. Models would need to be clearly labeled as AI-generated.

For exceptionally high-risk generative AI deemed to pose “systemic risks,” additional oversight rules would apply. This could include forming an industry council to monitor for harm, documenting any incidents, and coordinating with EU regulators.

Addressing harm while supporting innovation

For European officials, a key priority has been balancing the need to foster an indigenous AI industry against managing potential societal dangers from uncontrolled AI systems.

Some European nations like France and Germany initially opposed regulations that they felt could undermine domestic AI startups and research hubs. The final language appears to have assuaged those concerns.

With elections coming in June, officials have raced to finalize language and pass regulations before a new Commission and Parliament takes office. The technical specifics will still need to be refined in follow-up meetings between national representatives.

But the broad guardrails now appear set. And the world is watching what could become the template for AI governance for years to come.

The post EU poised to approve sweeping new AI regulations appeared first on ReadWrite.

Steve Hotelling, Apple’s senior vice president overseeing the development of touchscreen technology, health sensors, and biometric authentication for iPhone and other devices, is retiring after over 20 years with the company, according to a recent Bloomberg report. The departure comes at a critical juncture as Apple prepares to launch its first AR/VR headset and other products featuring innovative display and sensor systems that Hotelling spearheaded.

Known internally as the “father of the iPhone multi-touch screen”, Hotelling led teams responsible for some of Apple’s most complex integrated hardware and software. This includes the responsive and intuitive touchscreens that define the iPhone experience as well as new biometric sensors like Face ID. “No one was more brilliant than Steve,” said a longtime peer at Apple.

A valuable team member

With hundreds of patents to his name, Hotelling represented Apple in multiple high-stakes trials over iPhone intellectual property, including touchscreen patents that were part of its heated global legal battle against Samsung last decade. Most recently, he was a key witness in an ongoing patent lawsuit brought by medical device maker Masimo Corp. regarding Apple’s development of health sensors.

Hotelling’s responsibilities spanning camera systems, displays, haptics, augmented reality depth-sensing and more are now being handled by multiple VPs reporting to SVP Johny Srouji. This includes Alan Gilchrist overseeing camera hardware and Wei Chen managing certain display projects. However, sources say Hotelling’s departure leaves big shoes to fill when it comes to driving cross-disciplinary technological innovation.

The retirement also comes amid reported struggles for Apple’s hardware technologies group to ship other planned launches like in-house modem chips and microLED displays on schedule – key components towards reducing dependence on external suppliers. With the team racing to debut Apple’s first AR headset next year and integrate cutting-edge technologies into future products, some engineers may be feeling Hotelling’s absence more acutely in the months ahead.

The post Apple’s touchscreen and Face ID chief departs after over 20 years appeared first on ReadWrite.

Google has unveiled a major upgrade to its conversational AI chatbot Bard, now powering it with the company’s own new language model called Gemini, according to a recent The Verge report. With its first rollout in 170 countries today for English users, Google hopes this will make Bard a viable rival to ChatGPT and other leading chatbots built on GPT-3.5 and GPT-4 models.

Bard is now running on Gemini Pro, described as the “Goldilocks” version of Gemini that balances speed, efficiency, and capability for chatbot applications. It sits between Gemini Ultra, the largest and most advanced version for complex reasoning, and Gemini Nano, optimized for on-device tasks. Powered by Gemini Pro, Google believes Bard can now match or outdo what ChatGPT can offer across key areas like summarization, writing assistance, question answering, and more.

Multimodal strengths for images, audio, video

One distinct advantage of Gemini is its strength for multimodal applications with images, audio, and video. Built from the ground up to seamlessly integrate across modalities, Gemini enables more interactive features like getting AI feedback on photos of design prototypes or homework problems. This could allow for more visual interaction modes in the planned “Bard Advanced” release next year based on Gemini Ultra.

CEO Sundar Pichai suggests this upgrade represents a major quality leap for Bard, with improvements in understanding intent, factual accuracy, and overall response quality. But he simultaneously described it as just the start for Gemini models powering conversational AI. The goal is to keep rapidly iterating to achieve capabilities on par with OpenAI’s GPT models that dominate today.

The launch leaves Bard better equipped to challenge rivals in key chatbot performance benchmarks. But with multimodal and advanced Gemini versions still forthcoming, Google aims higher to make Bard AI competitive through continuous modeling breakthroughs. For them, today marks meaningful progress en route to that end goal.

The post Google upgrades Bard with new Gemini model to rival ChatGPT appeared first on ReadWrite.

Governments around the world, including allies of the United States, have been secretly surveilling smartphone users by exploiting push notifications from popular apps, according to information revealed this week by U.S. Senator Ron Wyden.

In a letter sent Wednesday to the Department of Justice, Senator Wyden warned that unidentified foreign governments have been demanding metadata related to app push notifications from tech giants Apple and Google in order to reveal private information about smartphone owners.

Push notifications, the audible “dings” or on-screen messages smartphone users receive from apps like email, messaging, and social media apps, provide insights into real-time app usage and activity. By obtaining metadata on push notification traffic from Apple and Google, who transmit notifications via their servers, government agencies can potentially identify anonymous app users and tie activity back to specific Apple and Google accounts and devices.

According to a recent Reuters report, sources familiar with the surveillance say that while details remain unclear, both foreign democracies allied with the U.S., as well as domestic agencies, have made requests for push notification metadata, likely to aid criminal or intelligence investigations by connecting online activities to real-world identities.

In his letter, Senator Wyden admonished the DOJ for hindering public knowledge of this surveillance method, which relies on the fact that most smartphone owners are unaware of how integrally involved Apple and Google are in transmitting app notifications. In response, Apple claimed they had been explicitly barred from publicly discussing demands for push notification data until now.

Both Apple and Google stated they are committed to keeping users informed going forward. However, the prevalence of apps using push notifications, from email and messaging to sports and breaking news, means a wide swath of smartphone activity worldwide may have already been caught up in surveillance efforts.

While the scale, duration, and details of the surveillance are still emerging, Senator Wyden’s revelations confirm that governments worldwide are actively exploiting common smartphone functions like notifications to secretly spy on mobile device owners without their knowledge or consent.

The post US Senator reveals allies are spying on smartphone users via Push Notifications appeared first on ReadWrite.

Apple is developing several new iPad and Mac models with plans to release them in early 2023, according to inside sources cited in a recent Bloomberg article. The upcoming product launches aim to boost declining sales amid a slowdown in consumer tech spending.

The new releases include updated versions of the iPad Air, iPad Pro, and MacBook Air. The iPad Air will now come in two sizes – 10.9 inches as currently offered, and a new 12.9-inch model matching the largest iPad Pro. The move gives consumers a bigger screen at a lower price point than the pricier Pro.

The new iPad Pros are slated to receive Apple’s high-end M3 chip introduced last fall, along with the company’s first OLED tablet displays. The organic light-emitting diode screens offer brighter colors and contrast compared to standard LCDs. New Apple Pencil and Magic Keyboard accessories are also in development.

On the Mac side, the new 13-inch and 15-inch MacBook Air models will feature the new M3 chip as their main upgrade. The processor promises faster performance and efficiency gains.

A response to slowing sales

iPad sales dropped 10% last quarter to $6.44 billion, while Mac revenue declined 34% to $7.61 billion. The products now jointly account for only 15% of Apple’s total sales – down significantly in recent years. Services have rapidly grown to represent a much larger business for Apple.

The company likely hopes its 2024 tablet and laptop upgrades will spur demand after what will be the first calendar year without any new iPad releases. The refreshed models align with Apple’s typical spring launch cycle.

Beyond early 2024, Apple is readying additional products like its Vision augmented reality headset, updated Apple Watches with blood pressure tracking, new iPhone models, and more. But in the nearer term, revived iPad and MacBook sales could boost Apple’s bottom line after recent consumer tech headwinds.

The post Apple preparing new iPad and Mac models to reinvigorate sales appeared first on ReadWrite.

Apple Inc. has quietly released new machine learning frameworks and model libraries, MLX and MLX Data, signaling a growing interest in pursuing generative AI applications, according to a recent The Verge report. The releases come as a surprise given Apple’s historically conservative approach to AI technology.

MLX is a machine learning framework optimized to run efficiently on Apple Silicon chips. It makes it easy for developers to build AI models that can run on MacBook and other Apple devices. MLX Data is a supplementary data loading framework.

According to Apple’s GitHub page, MLX draws inspiration from existing frameworks like PyTorch and Jax but is uniquely designed for seamless integration across Apple hardware, utilizing shared memory for faster model training. Developers have noted that MLX appears capable of training complex AI models comparable to Meta’s Llama and Stable Diffusion.

Will Apple pursue the ever-expanding generative AI space?

The launch suggests Apple may be poised to pursue the popular generative AI space, including text, image, and video generation, an area companies like Microsoft and Google have rapidly expanded into recently.

While Apple has utilized AI technology in its products for years, it has traditionally avoided labeling it as such, preferring terms like “machine learning.” The company also appears to have accelerated its AI development efforts this past September to evaluate foundational models for broader implementation.

Industry analysts theorize Apple may leverage its new frameworks to introduce innovative generative AI features across its services and devices — although the historically privacy-centric company will likely assess ethical implications closely first. Apple declined requests for further comment on its specific MLX, MLX Data, and generative AI plans.

Generative AI refers to a class of artificial intelligence capabilities focused on actually creating or “generating” new content rather than just analyzing data or responding to input. The most common types of generative AI generate synthetic text, images, audio, and video that closely mimic human-produced creative works.

Featured Image Credit: Photo by Armand Valendez; Pexels

The post New AI frameworks hint at Apple’s generative AI roadmap appeared first on ReadWrite.